CN107847746B

CN107847746B - Triggering atrial fibrillation electrodes in an implantable device

Info

Publication number: CN107847746B
Application number: CN201680044683.8A
Authority: CN
Inventors: 迪帕·马哈詹; 戴维·L·佩尔施巴谢; 基思·L·赫尔曼
Original assignee: Cardiac Pacemakers Inc
Current assignee: Cardiac Pacemakers Inc
Priority date: 2015-07-30
Filing date: 2016-06-07
Publication date: 2021-10-22
Anticipated expiration: 2036-06-07
Also published as: JP6653372B2; US10617320B2; WO2017019178A1; EP3328483B1; JP2018523518A; EP3328483A1; CN107847746A; US20170027462A1

Abstract

An apparatus comprising: a sensing circuit configured to generate a sensed physiological signal representative of cardiac activity of a subject; an arrhythmia detection circuit; control circuitry and memory. The arrhythmia detection circuit detects an onset of Atrial Fibrillation (AF) in the sensed cardiac signal using a first AF detection criterion and detects the onset of AF using a second AF detection criterion. The first AF detection standard has a higher sensitivity for AF detection than the second AF detection standard, and the second AF detection standard has a higher specificity for AF detection than the first AF detection standard. When an episode of AF is detected by both the first AF detection criterion and the second AF detection criterion, the control circuitry initiates storage of sampled values of a segment of the cardiac signal that includes the episode of AF.

Description

Triggering atrial fibrillation electrodes in an implantable device

Priority requirement

This application claims priority benefits under 35u.s.c. § 119(e) to U.S. provisional patent application serial No. 62/199,022 filed on 30/7/2015, which is incorporated herein by reference in its entirety.

Background

Portable (album) medical devices include Implantable Medical Devices (IMDs), wearable medical devices, handheld medical devices, and other medical devices. Some examples of IMDs include Cardiac Function Management (CFM) devices such as implantable pacemakers, Implantable Cardioverter Defibrillators (ICDs), subcutaneous implantable cardioverter defibrillators (S-ICDs), cardiac resynchronization therapy devices (CRTs), and devices including combinations of such capabilities. The device may be used to treat a patient or subject using electrotherapy or other therapies, or to assist a doctor or caregiver in patient diagnosis by internally monitoring the patient's condition.

Some implantable medical devices may be diagnostic-only devices, such as Implantable Loop Recorders (ILRs) and subcutaneous implantable heart failure monitors (SubQ HFMs). The apparatus may include electrodes in communication with one or more sense amplifiers to monitor heart electrical activity within the patient, or may include one or more sensors for monitoring one or more other internal patient parameters. The subcutaneous implantable device may include electrodes capable of sensing cardiac signals without directly contacting the patient's heart. Other examples of IMDs include implantable drug delivery systems or implantable devices with neurostimulation capabilities (e.g., vagus nerve stimulators, baroreflex stimulators, carotid sinus stimulators, spinal cord stimulators, deep brain stimulators, etc.).

Some examples of wearable medical devices include Wearable Cardioverter Defibrillators (WCDs) and wearable diagnostic devices (e.g., portable monitoring vests, holter monitors, cardiac event monitors, or mobile cardiac telemetry devices). The WCD may be a monitoring device that includes surface electrodes. The surface electrodes may be arranged to provide monitoring to provide a surface Electrocardiogram (ECG) and to deliver one or both of cardioversion and defibrillator shock therapy. In some examples, the wearable medical device may also include a monitoring patch worn by the patient, such as an adherable patch or may be included on an article of clothing worn by the patient.

Some examples of handheld medical devices include Personal Data Assistants (PDAs) and smart phones. The handheld device may be a diagnostic device that records an Electrocardiogram (ECG) or other physiological parameter while the device is stationary in a patient's hand or held on the patient's chest.

CFM devices are implantable, but may not include dedicated atrial sensing capability in some cases. In addition, some implantable, wearable, and handheld devices that are diagnostic only do not include dedicated atrial sensing capabilities. Patients with these types of devices may develop atrial arrhythmias, such as Atrial Fibrillation (AF). This is particularly true for heart failure patients who typically have a high incidence of AF. Knowing that a particular patient is experiencing AF may be useful to physicians and clinicians, either for diagnostic purposes or to tailor the performance of a medical device to the patient's needs to provide the most effective patient therapy.

Disclosure of Invention

For portable medical devices, it is desirable to properly detect and identify arrhythmias. This may help provide the most effective device-based therapy or non-device based therapy for the patient. The present subject matter relates to improving the detection of atrial fibrillation and recording the onset of fibrillation.

One example apparatus of the present subject matter may include: sensing circuitry, arrhythmia detection circuitry, memory, and control circuitry. The sensing circuitry is configured to generate a sensed cardiac signal representative of cardiac activity of the subject. The arrhythmia detection circuit is configured to detect an onset of Atrial Fibrillation (AF) in the sensed cardiac signal using a first AF detection criterion, and detect the onset of AF using a second AF detection criterion. The first AF detection criterion has a higher sensitivity for AF detection than the second AF detection criterion, and the second AF detection criterion has a higher specificity for AF detection than the first AF detection criterion. The control circuit is configured to initiate storage of sampled values of a segment of the cardiac signal including the episode of AF in the memory when the episode of AF is detected by both the first AF detection criteria and the second AF detection criteria.

This section is intended to provide a brief summary of the subject matter of the present patent application. And are not intended to provide an exclusive or exhaustive explanation of the invention. The detailed description is included to provide further information about the present patent application, such as discussing the dependent claims and the interrelationship of the dependent and independent claims in addition to the statements made in this section.

Drawings

In the drawings, which are not necessarily drawn to scale, like numerals may describe similar components in different views. Like numerals having different letter suffixes may represent different instances of similar components. By way of example, and not limitation, the figures generally illustrate various examples discussed in this document.

Fig. 1 is an illustration of an example of a portion of a medical device system including an IMD.

Fig. 2 and 3 are diagrams of further examples of IMDs.

Fig. 4 is an illustration of a portion of another example of a medical device system.

Fig. 5 shows a flow chart of an example of a method of operating a portable medical device.

Fig. 6 shows a block diagram of portions of an example of a portable medical device.

Fig. 7 shows a representation of sensed cardiac signals.

Fig. 8 shows an example of a sensed physiological signal with both normal sinus rhythm and atrial fibrillation.

Fig. 9 shows a graph of an example of a heart rate distribution for a normal sinus rhythm.

Fig. 10 shows a graph of an example of a heart rate distribution for atrial fibrillation patients.

Fig. 11 illustrates a method of triggering storage of a cardiac signal in response to detection of atrial fibrillation.

Fig. 12 shows an example of another method of triggering storage of a cardiac signal in response to detection of atrial fibrillation.

Fig. 13 illustrates an example of yet another method of triggering storage of a cardiac signal in response to detection of atrial fibrillation.

Fig. 14 shows an example of a menu of different options that may be used for atrial fibrillation detection.

Fig. 15 shows an example of yet another method of triggering storage of a cardiac signal in response to atrial fibrillation detection.

Detailed Description

The portable medical device may include one or more of the features, structures, methods, or combinations thereof described herein. For example, a cardiac monitor or cardiac stimulator may be implemented to include one or more of the advantageous features or processes described below. Such monitors, stimulators, or other portable devices are not intended to include all of the features described herein, but may be implemented to include selected features that provide unique structures or functions. Such devices may be implemented to provide various therapeutic or diagnostic functions.

Fig. 1 is an illustration of an example of a portion of a system 100, the system 100 including a portable medical device that is an IMD 105. Examples of IMDs 105 include, but are not limited to, pacemakers, cardioverters, defibrillators, and other cardiac monitoring and therapy delivery devices (including cardiac devices that contain or work in conjunction with one or more neurostimulation devices, drugs, drug delivery systems, or other therapies). In an example, the system 100 shown is used to treat cardiac arrhythmias. The IMD 105 generally includes an electronics unit coupled to the heart of a patient or subject via one or more cardiac leads 115. The electronics unit of the IMD 105 generally includes components enclosed in a sealed enclosure sometimes referred to as a cartridge or "can". The system 100 also typically includes an IMD programmer or other external system 190, the other external system 190 communicating one or more wireless signals 185 with the IMD 105, such as by using Radio Frequency (RF) or by one or more other telemetry methods.

The example shown includes a Right Ventricular (RV) lead 115 having a proximal portion and a distal portion. The proximal end is coupled to a header connector 107. The distal portion is configured to be placed in the RV. The RV lead 115 can include one or more of a proximal defibrillation electrode 116, a distal defibrillation electrode 118 (e.g., RV coil), an RV tip electrode 120A, and an RV ring electrode 120B. Defibrillation electrode 116 is typically incorporated into a lead body, such as in a location suitable for supraventricular placement in the superior vena cava (e.g., SVC coil). In some examples, RV lead 115 includes a ring electrode 132 (e.g., an SVC ring) near proximal defibrillation electrode 116. Defibrillation electrode 118 is incorporated into the lead body near the distal end, such as placed in the RV.

RV electrodes

120A and 120B may form a bipolar electrode pair and are typically incorporated into the lead body at the distal end of the lead.

Electrodes

116, 118, 120A, and 120B are each electrically coupled to IMD 105, such as by one or more conductors extending within the lead body. The proximal defibrillation electrode 116, the distal defibrillation electrode 118, or an electrode formed on the can of the IMD 105 allows for the delivery of cardioversion or defibrillation pulses to the heart. RV tip electrode 120A, RV ring electrode 120B or an electrode formed on the can of IMD 105 allows sensing of RV electrogram signals indicative of RV depolarization and delivery of RV pacing pulses. The IMD 105 includes sense amplifier circuitry for providing amplification or filtering of the sensed signals. Sensing and pacing allows the IMD 105 to adjust the timing of ventricular contractions.

Some IMDs may not include any electrodes for sensing electrical activity in the atrium, such as shown in fig. 1. For example, the IMD 105 may be an ICD with single ventricular chamber sensing. ICDs may include electrodes attached to a single ventricular lead and use intracardiac signals sensed by the ventricular electrodes for arrhythmia detection and discrimination (e.g., by rate sensing and/or depolarization signal morphology analysis).

An IMD may be a diagnostic-only device and does not provide electrical therapy to the patient. Such devices may include RV tip electrode 120A, RV ring electrode 120B or a combination of electrodes formed on the can of IMD 105 to allow sensing of ventricular depolarizations. Note that the particular arrangement of leads and electrodes shown as the illustrated example of fig. 1 is intended to be non-limiting.

Fig. 2 is an illustration of another example of a portion of a system 200 that includes an S-ICD 205. S-ICD205 is subcutaneously implantable and includes lead 215. Lead 215 is also implanted subcutaneously and the proximal end of lead 215 is coupled to header connector 207. Lead 215 may include

electrodes

220A and 220B for sensing ventricular depolarization (e.g., using far field sensing), but in the example shown, the lead does not include any electrodes that directly contact the heart. Lead 215 includes defibrillation electrode 218, which may be a coil electrode. S-ICD205 may provide one or more of cardioversion therapy and defibrillation high-energy shock therapy to the heart using defibrillation electrode 218 and an electrode formed on the can of S-ICD 205. In some examples, S-ICD205 may also provide pacing pulses for anti-tachycardia therapy or bradycardia therapy. Note that direct atrial sensing is not provided in the electrode arrangement, but the

electrodes

220A and 220B allow sensing of far-field ventricular electrogram signals.

Fig. 3 is an illustration of an example of a leadless IMD. In the example shown, the IMD is a leadless pacemaker 305. The leadless pacemaker 305 is shown positioned at the endocardium within a ventricular cavity, but the leadless pacemaker 305 may be positioned at other locations of the heart. The leadless pacemaker 305 example has a cylindrical or bullet-shaped housing and may include one or more electrodes disposed along the cylindrical housing to sense electrical signals of the heart and/or provide electrical stimulation for pacing the heart. One or more electrodes may be used for communication. The leadless pacemaker 305 may include a mechanism 330 to secure the pacemaker to the myocardium. Examples of the securing mechanism members may include one or more tines, one or more barbed tines, and one or more helical securing mechanism members. The electrodes shown in the examples for device placement may not provide direct atrial sensing, but the electrodes may provide RV electrogram signals.

Other examples of IMDs include Implantable Loop Recorders (ILRs), leadless diagnostic devices in the heart, and neurostimulators (including but not limited to vagal nerve stimulators, baroreceptor stimulators, and spinal cord stimulators), or other IMDs. These types of devices may not include electrodes positioned in the atrium.

Fig. 4 is an illustration of portions of another example of a medical device system 400. The system 400 may include one or more portable medical devices, such as a conventional implantable or subcutaneous implantable medical device 405, a wearable medical device 410, or a handheld medical device 403. One or more medical devices may include communication circuitry (e.g., telemetry circuitry) to communicate an indication of AF to communication system 407. The communication system 407 may include an external communication device 412 and a remote system 414, the remote system 414 communicating with the external communication device 412 via a network 418 (e.g., the internet, a proprietary computer network, or a cellular telephone network). The remote system 414 may include a server 416 remote from the external communication device 412 and the subject to perform patient management functions. The external communication device 412 may include a programmer for programming treatment parameters of a device-based therapy provided by the implantable medical device. One or both of the external communication device 412 and the remote system 414 may include a display for presenting an indication of AF to a user, such as a clinician.

Fig. 5 is a flow chart of an example of a method 500 of operating a portable medical device. Method 500 provides for detecting AF using a portable medical device even though the portable medical device may not include electrodes and sensing circuitry for performing direct atrial sensing and recording of detected AF episodes.

At 505, cardiac electrical signals representative of heart activity of a subject are sensed using a portable medical device. In some examples, the cardiac signal may include information corresponding to a ventricular depolarization interval (or V-V interval) of the subject.

At 510, the episode of AF is detected in the sensed cardiac signal using a first AF detection criterion, and the episode of AF is detected or re-detected using a second AF detection criterion at 515. Thus, AF detection may be twofold. The first AF detection standard may have a higher sensitivity to AF detection than the second AF detection standard, and the second AF detection standard has a higher specificity to AF detection than the first AF detection standard. Sensitivity refers to the ability of the detection scheme of the device to effectively detect abnormal heart rhythms (e.g., AF) that the device may treat. Specificity refers to the ability of the detection scheme of the device to correctly identify heart rhythms for which the device is not intended to treat (e.g., normal rhythms, non-AF arrhythmias, or noise incorrectly identified as arrhythmias).

In some examples, the second AF detection criteria is a different detection method than the first AF detection criteria. In some examples, the second AF detection criterion is the same detection method as the first AF detection criterion, but the threshold for AF detection is sufficiently different to increase the detection specificity of the second AF detection criterion relative to the first AF detection criterion.

At 520, when an episode of AF is detected by both the first AF detection criteria and the second AF detection criteria, sampled values are stored as a segment of the cardiac signal that includes the detected episode of AF. This allows for the storage of AF electrograms that can be later uploaded for evaluation by a clinician.

Fig. 6 shows a block diagram of portions of an example of a portable medical device. The apparatus 605 includes a sensing circuit 610, a control circuit 615, an arrhythmia detection circuit 620, and may include a memory 625. The sensing circuit 610 may generate a sensed cardiac signal representative of cardiac activity of the subject. In certain examples, sensing circuit 610 may be electrically coupled to one or more implantable electrodes included in a lead arranged for placement in a heart chamber. In certain examples, sensing circuit 610 may be electrically coupled to one or more implantable electrodes included in a leadless implantable medical device. In certain examples, sensing circuitry 610 may be electrically coupled to one or more implantable electrodes configured to sense cardiac signals without direct cardiac contact with the subject (e.g., subcutaneous implantable electrodes). In certain examples, the sensing circuit 610, the control circuit 615, the arrhythmia detection circuit 620, and the memory 625 are included in a wearable device or a handheld device. In a variant, the memory may be included in a separate device, or may be a central memory located in a network "cloud".

The control circuit 615 may include a microprocessor, digital signal processor, Application Specific Integrated Circuit (ASIC), or other type of processor that interprets and executes instructions included in software or firmware. The memory 625 may be integrated into the control circuit 615 or separate from the control circuit 615. The arrhythmia detection circuit 620 may also be integrated into the control circuit 615 or may be separate from the control circuit 615. In some examples, the sensing circuit 610 is included in a first device and the arrhythmia detection circuit and the control circuit are included in a second, separate device. In certain examples, the first device is implantable and the second device is external.

Arrhythmia detection circuit 620 detects the onset of AF in the sensed cardiac signal using a first AF detection criterion and detects the onset of AF using a second AF detection criterion. In response to AF detection by both the first AF detection criteria and the second AF detection criteria, the control circuit 615 initiates or triggers storage of sampled values of a segment of the cardiac signal (e.g., in the memory 625 or a different memory) that includes an AF episode.

According to some examples, the first AF detection criterion is a measure of ventricular depolarization (V-V) interval dispersion. Fig. 7 shows a representation of a sensed cardiac signal 705. The signal is shown with a plurality of R-waves 710. The V-V interval may be determined as an interval between R-waves. RR1 in the figure refers to the first interval between the first two R-waves; RR2 is the second interval between the second R-wave and the third R-wave, and so on. The difference between the V-V intervals is called Δ RR_1,2(e.g., difference between RR2 and RR 1), Δ RR_2,3And so on.

Fig. 8 shows an example of a sensed physiological signal having a first region 805 corresponding to NSR and a second region 810 corresponding to AF. In the NSR region, the V-V spacing will be more regular and the difference in V-V spacing will be smaller. In the AF region, the V-V intervals will be more dispersed, and the value of the difference in V-V intervals will vary more than the NSR.

In some examples, arrhythmia detection circuit 620 uses the sensed physiological signal to determine a ventricular depolarization (V-V) interval and monitors information corresponding to the V-V interval. Arrhythmia detection circuit 620 may include a peak detector circuit to detect R-waves in the sensed physiological signal to determine the V-V interval. Arrhythmia detection circuit 620 may sample the V-V intervals and store the samples in device memory 625 or a different memory. Arrhythmia detection circuit 620 may determine a difference between the V-V intervals and use the determined V-V interval difference to determine a measure of V-V interval dispersion.

In some examples, the measure of the spread between V-V intervals comprises a determined variance of the determined interval difference. As a first AF detection criterion, arrhythmia detection circuit 620 may compare the determined variance to a specified variance threshold and generate an indication of AF when the determined variance satisfies the specified variance threshold.

Other measures of ventricular interval dispersion may be used as the first AF detection criterion. In some examples, arrhythmia detection circuit 620 determines a difference in V-V intervals and classifies the interval difference as one of stable, unstable, or unstable and random. Interval classification can be used to determine V-V interval dispersion.

In some variations, the intervals are classified as stable bins (bins), unstable bins, or unstable-random bins. An interval difference may be classified as stable when it is less than a specified threshold difference value from the immediately preceding interval difference; an interval difference may be classified as unstable when it is greater than a specified threshold difference value from the immediately preceding interval difference; and when the magnitude of the interval difference is greater than a specified threshold difference value of the immediately preceding interval difference and the interval difference is negative (which satisfies a specified negative threshold), the interval difference may be classified as unstable-random.

In some examples, the threshold difference value is a value corresponding to less than 10bpm of the rate difference between the two intervals. Thus, if RR2 in FIG. 7 is 1000ms corresponding to 60bpm and RR1 is 857ms corresponding to 70bpm, the difference in separation Δ RR_1,2Are binned as stable. If RR1 is less than 857ms, the interval difference is binned as unstable. If RR2 is less than 857ms and RR1 equals 1000ms, the interval difference Δ RR_1,2Binned as unstable-random. In some examples, the interval differences are only considered for binning if the intervals used (e.g., intervals RR1 and RR2) are included in a trio of three ventricular beats that are longer than the specified minimum interval (e.g., interval 324ms corresponding to heart rate 185 bpm).

In the example of fig. 8, more V-V spacing differences will be stable in the NSR region. In the AF area, the number of unstable V-V interval differences and unstable-random V-V interval differences will increase relative to the number of stable V-V interval differences. The arrhythmia detection circuit 620 may determine a first metric of ventricular interval dispersion using a number of the plurality of stable interval differences and a number of the plurality of unstable interval differences. The first metric may include a first ratio determined using a number of the plurality of stable interval differences and a number of the plurality of unstable interval differences (e.g., the first ratio ═ unstable/stable).

Returning to the apparatus 605 of fig. 6, the arrhythmia detection circuit 620 may determine a second measure of ventricular interval dispersion using the determined portion of the unstable-random interval difference. The second metric may include a second ratio determined using a plurality of unstable-random interval differences and including a sum of the number of stable interval differences and the number of unstable interval differences (e.g., the second ratio ═ unstable-random)/(stable + unstable)).

In the case where the first and second metrics are the ratio, the value of the first ratio will increase in the presence of AF because the number of V-V intervals classified as unstable will increase. In the presence of AF, the value of the second ratio will tend to increase, since the number of V-V interval differences classified as unstable-random will increase.

As a first AF detection criterion, arrhythmia detection circuit 620 may compare the determined first ratio to a specified first ratio threshold (e.g., a ratio value of 3) and compare the determined second ratio to a specified second ratio threshold (e.g., a ratio value of 0.06 or 6%). The arrhythmia detection circuit 620 generates an indication of AF when the first ratio satisfies a specified first ratio threshold and the determined second ratio satisfies a specified second ratio threshold.

As a second AF detection criterion, in some examples, arrhythmia detection circuit 620 uses the sampled VV interval values to determine a VV interval distribution and determines a Heart Rate Density Index (HRDI) as a fraction of the sampled VV interval values corresponding to the VV intervals that occur most frequently in the distribution.

Fig. 9 shows a graph of an example of a heart rate distribution of a Normal Sinus Rhythm (NSR). Alternatively, the distribution may be a V-V interval distribution. Most of the samples of the distribution lie between about 50bpm and 90 bpm. In some variations, the HRDI may be expressed as a fraction (e.g., a percentage) of the interval. In the example of fig. 9, HRDI is 81% of the heart rate mode corresponding to 60 bpm. Fig. 10 shows a graph of an example of a heart rate distribution of a patient in AF. It can be seen that heart rate is more irregular in AF than in NSR. In the example of FIG. 10, HRDI is approximately 23%.

Returning to the apparatus 605 of fig. 6, the arrhythmia detection circuit 620 compares the HRDI to a specified HRDI threshold and generates an indication of AF when the determined HRDI satisfies the HRDI threshold. In the example of fig. 8, arrhythmia detection circuit 620 may generate an indication of AF as a second AF detection criterion when HRDI is less than 25%.

As explained previously, the first AF detection standard may have a higher sensitivity to AF detection than the second AF detection standard, and the second AF detection standard may have a higher specificity to AF detection than the first AF detection standard. The difference in sensitivity and specificity can be implemented by adjusting the threshold of the first AF detection criterion and the second AF detection criterion. For example, the specified V-V dispersion threshold of the first AF detection criterion may be lowered to include more candidate rhythms as AF, and the HRDI of the second AF detection criterion may be lowered to make the second AF detection more difficult to satisfy.

Other AF detection methods may be used for one or both of the first AF detection criterion and the second AF detection criterion. In some examples, the first AF detection criteria may include determining a heart rate pattern. Returning to the heart rate distribution of fig. 9, the pattern of the distribution is about 60bpm, since that is the most frequently occurring heart rate in the distribution. In the example of fig. 10, the heart rate mode is 90 bpm.

Returning to apparatus 605 of fig. 6, arrhythmia detection circuit 620 may determine the heart rate mode as the heart rate corresponding to the V-V interval value having the most samples in the V-V interval distribution. Arrhythmia detection circuit 620 compares the heart rate pattern to a specified heart rate pattern threshold and generates an indication of AF when the heart rate pattern meets the specified heart rate pattern threshold. For example, the arrhythmia detection circuit 620 may generate an indication of AF when the heart rate mode is at intervals of 600 milliseconds (600ms) or less (corresponding to a heart rate greater than or equal to 100 bpm).

In some examples, one or both of the first AF detection criteria and the second AF detection criteria use morphology of the sensed cardiac signal to detect AF. Arrhythmia detection circuit 620 may include a scoring module 630 that determines a score associated with a correlation of the morphology of the sensed cardiac signal with a morphology of a template signal that identifies AF. An example of an associated score is a Feature Correlation Coefficient (FCC). The FCC may provide an indication of the degree of similarity between the shape of the sensed electrogram and the shape of the template electrogram signal representing AF. The template may be recorded for a particular subject or may be created based on a patient population. Methods for calculating a relevance score can be found in U.S. patent No.7,904,142 entitled "Self-Adjusting ECG morphology correction Threshold" filed on 16.5.2007, which is incorporated herein by reference in its entirety.

In some examples, arrhythmia detection circuit 620 applies at least one of the first AF detection criteria and the second AF detection criteria to detect the onset of AF using the determined score. AF may be detected when the determined score satisfies a specified threshold score. By adjusting the threshold score, the detection of AF can be adjusted to be sensitive or less sensitive.

The values of V-V interval dispersion, heart rate mode, HRDI, and morphology score can only be measured using ventricular sensing. As such, AF may be detected in a portable medical device without including dedicated atrial sensing. Other criteria may be used if other sensing circuits are available.

In some examples, the apparatus 605 of fig. 6 includes a physiological sensor circuit 635. The physiological sensor circuit 635 generates a physiological signal including physiological information of the subject. During AF, the performance of the subject's hemodynamic system may be degraded. Degraded performance may be reflected in the physiological signal.

An example of the physiological sensor circuit 635 is a Pulmonary Arterial Pressure (PAP) sensor circuit. The PAP sensor circuit may be implanted in the pulmonary artery to sense the PAP signal. An example of an implantable PAP sensor is described in U.S. Pat. No.7,566,308 entitled "Method and Apparatus for Pulmonary electrode Pressure Signal Isolation" filed on 13/10/2005, the entire contents of which are incorporated herein by reference. The onset of AF may result in a decrease in PAP in the subject as reflected in the sensed PAP signal. In some examples, the arrhythmia detection circuit 620 applies the second AF detection criteria to the sensed PAP signal to confirm AF detection using the first AF detection criteria. In certain examples, the second AF detection criteria generates an indication of AF when the PAP reflected in the PAP signal decreases below a specified PAP threshold. The first AF detection criteria may be applied to the sensed electrogram signal as described previously herein, or may be criteria applied to the PAP signal, with detection by the first criteria being more sensitive than detection by the second criteria.

Another example of the physiological sensor circuit 635 is a heart sound sensor circuit. Heart sounds are associated with the activity of the heart from the patient and the mechanical vibrations of blood flowing through the heart. The heart sounds recur every cardiac cycle and are separated and classified according to the activity associated with the vibrations. The first heart sound (S1) is a vibrating sound produced by the heart during tightening of the mitral valve. The second heart sound (S2) marks the beginning of diastole. The third heart sound (S3) and the fourth heart sound (S4) are related to the filling pressure of the left ventricle during diastole.

The physiological signal may be a heart sound signal representing one or more heart sounds produced by the heart sound sensor circuit. Examples of heart sound sensors include accelerometers or microphones. Methods for measuring heart sounds can be found in U.S. patent No.7,115,096 entitled "Method and Apparatus for Monitoring of dental Hemodynamics", filed on 30.12.2002, the entire contents of which are incorporated herein by reference.

The onset of AF may result in a change in a measurable heart sound parameter in the heart sound signal. For example, during an AF episode, the amplitude of the S1 or S2 heart sounds may decrease. In some examples, arrhythmia detection circuit 620 applies the second AF detection criteria to the sensed heart sound signal to confirm AF detection using the first AF detection criteria. The first AF detection criterion may be applied to the sensed electrogram signal as described herein before, or may be a criterion applied to the heart sound signal, the detection by the first criterion being more sensitive than the detection by the second criterion. The first AF detection criterion may be applied to a different heart sound parameter than the second AF detection criterion.

As previously explained, in response to AF detection by both the first AF detection criteria and the second AF detection criteria, the control circuit 615 initiates storage in the memory 625 of sampled values (e.g., an electrogram) of a segment of the cardiac signal that includes an episode of AF.

FIG. 11 illustrates a method of triggering storage of a cardiac signal in response to detecting AF. Arrhythmia detection circuit 620 applies a first detection criterion during the main AF detection window 1105. In response to detecting AF using the first AF detection criteria, the control circuit 615 triggers storage of a sampled value of the cardiac signal.

The stored samples are represented in fig. 11 as AF confirmation window 1110. Arrhythmia detection circuit 620 applies a second AF detection criterion to the stored samples of the cardiac signal. If the arrhythmia detection circuit 620 also detects the onset of AF using the second AF detection criteria, the arrhythmia detection circuit 620 generates an indication of the detection of AF. The indication may be a signal that is communicated to the control circuit 615, and the control circuit 615 continues to store sample values of the cardiac signal in response to the generated AF indication. The samples stored in the AF confirmation window and subsequent samples may be stored in the memory 625. This is useful for uploading cardiac signals with confirmation of AF episodes for analysis. If AF is not detected by the second AF detection criteria during the AF confirmation window, the control circuit 615 can terminate storage of the cardiac signal and return to searching for the cardiac signal for AF using the first AF detection criteria.

Fig. 12 shows an example of another method of triggering storage of a cardiac signal in response to detecting AF. The memory 625 includes a first start buffer 640 and a second event memory buffer 645. The buffers may be in separate memories or may be in different areas of the same memory. The control circuit 615 may initiate storage of sample values of the sensed cardiac signal in a start buffer, which may be used as a main AF detection window. Arrhythmia detection circuit 620 may apply both the first AF detection criterion and the second AF detection criterion to a segment of the cardiac signal stored in start buffer 640.

If an AF event is not detected, the control circuit 615 can initiate rewriting the start buffer. If AF is detected using the first and second detection criteria, the control circuit 615 may initiate moving the start buffer contents to the event memory buffer 645 and storing the values of subsequent samples of the sensed cardiac signal. Event storage buffer 645 may include representations of both the start of AF and the confirmation of AF.

In some examples, if the control circuit 615 determines that sample values for a segment of the cardiac signal have not been stored for a specified period of time because AF cannot be detected by both the first criterion and the second criterion, the control circuit can initiate storing in memory sample values for a segment of the cardiac signal that includes an AF episode when the episode of AF is detected by only the first AF detection criterion. This provides for storing an electrogram showing AF detection by the first AF detection criterion and non-detection by the second AF detection criterion. This may be useful for detection that allows the electrogram to be viewed to adjust one or both AF detection criteria.

In some examples, arrhythmia detection circuit 620 applies a first AF detection criterion to the contents of start buffer 640. If an AF event is detected, the contents of the start buffer are moved to an event store buffer. Arrhythmia detection circuit 620 applies a second AF detection criterion to the contents of the event memory buffer to confirm AF.

Fig. 13 shows an example of another method of triggering storage of a cardiac signal in response to detecting AF. Arrhythmia detection circuit 620 applies a first AF detection criterion during an initial main AF detection window 1305. The first AF detection criteria may be applied to the sensed cardiac signal in real-time or may be applied to sample values of the sensed cardiac signal stored in the start buffer 640. When AF is detected using the first AF detection criterion, the control circuit 615 initiates storage of the sampled value of the cardiac signal. In certain variations, the control circuit 615 initiates storage of the sample values in the event storage buffer 645.

Arrhythmia detection circuit 620 applies a second AF detection criterion to the stored sampled values of the sensed cardiac signal. The second AF detection criteria are applied to the second detection window 1310. The second detection window may have a duration that is shorter than a duration of the main detection window. When AF is detected using the second AF detection criterion, the control circuit 615 continues to store sampled values of the cardiac signal. When no AF is detected (e.g., at window 1315) using the second AF detection criteria, the control circuit 615 terminates storing the sample values of the cardiac signal.

The portable medical device may use different combinations of the described examples of AF detection. A medical device system (e.g., the system of fig. 4) may be used to present a menu to a user (e.g., using a graphical user interface or GUI) to select from available AF detection options.

Fig. 14 shows examples of different options that can be used for AF detection. These options may be available to the user (e.g., using a GUI) or may be preprogrammed into the medical device system. Option 1405 shows the case where the ambulatory medical device is configured with a first AF detection criterion that uses a different detection method than a second AF detection criterion. The first and second AF detection criteria (represented as

methods

1 and 2 in fig. 14) may include any combination of detection methods previously described herein, such as VV interval dispersion, HRDI, heart rate pattern, heart sounds, PAP, and morphological correlation. Option 1410 shows a case where the first and second AF detection criteria use the same AF detection method (denoted as method 2), but different thresholds (denoted as thresholds a, b) are used for AF detection.

Option 1415 shows a case where different combinations of AF detection criteria are used to detect AF. In some examples, AF detection may involve two-layer detection. A combination of the two detection methods is used as the first layer. For example, the two detection methods may include V-V spacing dispersion and HRDI. The second layer detection uses a combination of HRDI and heart sound measurements. In some examples, AF detection involves only one layer of detection, and AF may be detected when AF is detected using either of the first combination or the second combination criteria. When AF is detected using one or both of the first and second layer AF detections, the control circuitry may begin to initiate storing sample values of the cardiac signal.

Option 1420 shows the case where six AF detection criteria are selectable. The portable medical device may be programmed to detect AF and initiate storage of a sampled value of the cardiac signal when any combination of two or more detection criteria are met. Option 1425 shows a case where different AF detection criteria may be enabled, and a case where different detection thresholds for the criteria may also be enabled. AF may be detected by the portable medical device when it is automatically determined that any combination of AF detection criteria and detection thresholds have been met.

In some examples, the user selects different AF detection criteria and also selects different detection thresholds for the criteria. Providing a medical device or medical device system (providing different selectable combination criteria for AF detection) allows a clinician to adjust the triggering of an AF electrogram for an individual patient to the individual needs of the patient. This may result in storing the AF electrogram for later upload and evaluation by a clinician most relevant to the patient's condition.

Fig. 15 shows another example of another method of triggering storage of cardiac signals in response to detection of AF. In this example, the arrhythmia detection circuit 620 applies the same AF detection criteria during the first AF detection window 1505 and the second continuous AF detection window 1510. The AF detection criteria may be any detection method previously described herein, such as VV interval dispersion, HRDI, heart rate pattern, heart sounds, PAP, and morphology correlation. In response to detecting AF during the first and second consecutive detection windows, the control circuit 615 triggers storage of sample values of the sensed cardiac signal.

Returning to fig. 6, as explained previously, the device 605 may be a diagnostic-only device. In some examples, the device 605 provides therapy to the subject. The apparatus may include a therapy circuit 650, which may be electrically coupled to the electrodes to provide anti-arrhythmic therapy to the subject. The control circuit 615 initiates delivery of anti-arrhythmia therapy in response to the indication of AF generated by the arrhythmia detection circuit 620.

Additional description and examples

Example 1 includes the following subject matter (such as a device), including: sensing circuitry configured to generate a sensed cardiac signal representative of cardiac activity of a subject; an arrhythmia detection circuit coupled to the sensing circuit; a memory and a control circuit. The arrhythmia detection circuit is configured to: detecting an onset of Atrial Fibrillation (AF) in the sensed cardiac signal using a first AF detection criterion; and detecting an AF episode using a second AF detection criterion. The first AF detection standard has a higher sensitivity for AF detection than the second AF detection standard, and the second AF detection standard has a higher specificity for AF detection than the first AF detection standard. The control circuit is configured to trigger storage of sampled values of a segment of the cardiac signal comprising the episode of AF when the episode of AF is detected by both the first AF detection criterion and the second AF detection criterion.

In example 2, the subject matter of example 1 optionally includes: an arrhythmia detection circuit configured to: monitoring information corresponding to ventricular depolarization (VV) intervals, determining a VV interval distribution using the sampled VV interval values, and further configured to determine a Heart Rate Density Index (HRDI) as a fraction of the sampled VV interval values corresponding to the VV intervals that occur most frequently in the distribution, comparing the HRDI to a specified HRDI threshold, and generating an indication of AF when the determined HRDI satisfies the HRDI threshold, in accordance with a second AF detection criterion.

In example 3, the subject matter of one or both of examples 1 and 2 optionally includes arrhythmia detection circuitry configured, in accordance with a first AF detection criterion, to determine a heart rate mode as corresponding to a heart rhythm having a VV interval value of a maximum number of samples in a VV interval distribution, compare the heart rate mode to a specified heart rate mode threshold, and generate an indication of AF when the heart rate mode satisfies the specified heart rate mode threshold.

In example 4, the subject matter of one or any combination of examples 1-3 optionally includes arrhythmia detection circuitry configured, in accordance with a first AF detection criterion, to: determining a difference between the VV intervals, and determining a measurement of VV interval dispersion using the determined VV interval difference; the measurement of the V-V interval dispersion is compared to a specified dispersion threshold, and an indication of AF is generated when the determined measurement of the V-V dispersion meets the specified dispersion threshold.

In example 5, the subject matter of one or any combination of examples 1-4 optionally includes a physiological sensor circuit configured to generate a physiological signal including physiological information of the subject, wherein the arrhythmia detection circuit is configured to apply second AF detection criteria to the sensed physiological signal to detect an AF episode.

In example 6, the subject matter of example 5 optionally includes the physiological sensor circuit comprising at least one of a heart sound sensor circuit and a pulmonary artery pressure sensor circuit.

In example 7, the subject matter of one or any combination of examples 1-6 optionally includes a scoring module configured to determine a score associated with a correlation of morphology of the sensed cardiac signal to morphology of a template signal representative of AF, and wherein the arrhythmia detection circuit is configured to apply at least one of the first AF detection criteria and the second AF detection criteria to detect an onset of AF using the determined score.

In example 8, the subject matter of one or any combination of examples 1-7 optionally includes a memory including a first start buffer and a second event store buffer, and wherein the control circuitry is configured to: storing sample values of the sensed cardiac signal in a start buffer and overwriting sample values previously stored in the start buffer; and in response to AF detection by the first and second AF detection criteria, initiating storage of sample values stored in the start buffer in the event memory and subsequent sample values of the sensed cardiac signal in the event memory buffer.

In example 9, the subject matter of one or any combination of examples 1-8 optionally includes control circuitry configured to store, in the memory, the sampled value of the cardiac signal in response to detecting the onset of AF using the first AF detection criteria, wherein the arrhythmia detection circuitry is configured to detect the onset of AF using the second AF detection criteria using the stored sampled value of the cardiac signal and generate an indication of detection of AF, and wherein the control circuitry is further configured to continue to store the sampled value of the cardiac signal in response to the generated indication of AF.

In example 10, the subject matter of one or any combination of examples 1-9 optionally includes a control circuit configured to store sample values of a first segment of cardiac signal in a memory for a first duration in response to the arrhythmia detection circuit detecting the onset of AF using a first AF detection criterion. The arrhythmia detection circuit is configured to apply a second AF detection criterion to the stored values of the first cardiac signal segment, and the control circuit is configured to store sampled values of one or more cardiac signal segments having a second duration that is less than the first duration when AF is detected by both the first AF detection criterion and the second AF detection criterion. The arrhythmia detection circuit is optionally configured to apply a second AF detection criterion to the one or more cardiac signal segments having the second duration, and the control circuit is optionally configured to continue storing sample values of the cardiac signal segments of the second duration while the arrhythmia detection circuit continues to detect AF in the one or more cardiac signal segments having the second duration.

In example 11, the subject matter of one or any combination of examples 1-10 optionally includes control circuitry configured to determine that sample values of a segment of the cardiac signal have not been stored for a specified period of time, and in response to the determination, initiate storage of sample values of a segment of the cardiac signal including the episode of AF in the memory when the episode of AF is detected by only the first AF detection criteria.

In example 12, the subject matter of one or any combination of examples 1-11 optionally includes: a therapy circuit configured to be coupled to the electrode to provide anti-arrhythmic cardiac therapy to the subject; and a control circuit configured to initiate delivery of anti-arrhythmic therapy in response to the generated AF indication.

In example 13, the subject matter of one or any combination of examples 1-12 optionally includes sensing circuitry configured to be coupled to: at least one of an implantable electrode configured to be placed in a heart chamber or a subcutaneous implantable electrode configured to sense cardiac signals without direct cardiac contact with the subject.

In example 14, the subject matter of one or any combination of examples 1-13 optionally includes the sensing circuit and the arrhythmia detection circuit is included in a wearable device or a handheld device.

Example 15 optionally includes a subject matter (such as a system), or may optionally be combined with the subject matter of one or any combination of examples 1-14 to include a subject matter comprising: configured to sense cardiac electrical signals representative of cardiac activity of the subject; and a second means configured for detecting an onset of Atrial Fibrillation (AF) in the sensed cardiac signal using the first AF detection criteria; detecting the onset of AF using a second AF detection criterion, wherein the first AF detection criterion has a higher sensitivity to AF detection than the second AF detection criterion, and the second AF detection criterion has a higher specificity to AF detection than the first AF detection criterion; and storing sampled values of a segment of cardiac signal that includes the AF episode when the episode of AF is detected by both the first AF detection criteria and the second AF detection criteria.

In example 16, the subject matter of example 15 optionally includes the second apparatus being configured to sample a value of a ventricular depolarization (V-V) interval. The second AF detection criteria optionally include: determining a V-V interval distribution using the sampled V-V interval values; determining a Heart Rate Density Index (HRDI) as a portion of the sampled V-V interval values corresponding to the V-V intervals that occur most frequently in the distribution; comparing the HRDI to a specified HRDI threshold; and generating an indication of AF when the determined HRDI satisfies the HRDI threshold.

In example 17, the subject matter of one or both of examples 15 and 16 optionally includes a second apparatus configured to sample values of a ventricular depolarization (V-V) interval. The first AF detection criteria optionally include: determining a difference between the monitored V-V intervals; determining a measurement result of the V-V interval dispersion by using the determined V-V interval difference; comparing the measurement result of the V-V interval dispersion with a specified dispersion threshold value; and generating an indication of AF when the determined measure of V-V dispersion satisfies a specified dispersion threshold.

Example 18 includes a subject matter (such as an apparatus), or may optionally be combined with the subject matter of one or any combination of examples 1-17 to include a subject matter comprising: sensing circuitry configured to generate a sensed cardiac signal representative of cardiac activity of a subject; arrhythmia detection circuitry configured to detect an onset of Atrial Fibrillation (AF) in the sensed cardiac signal using AF detection criteria; a memory; and a control circuit. The control circuit is configured to:

initiating that a cardiac signal is to be sensed during a specified detection duration as an AF detection window; initiating sensing a cardiac signal during a first AF detection window and a second continuous AF detection window; and triggering storage of sample values of the sensed cardiac signal in response to detecting AF during both the first detection window and the second detection window.

In example 19, the subject matter of example 18 optionally includes: an arrhythmia detection circuit configured to: monitoring information corresponding to ventricular depolarization (V-V) intervals; determining a V-V interval distribution using the sampled V-V interval values; determining a Heart Rate Density Index (HRDI) as a portion of the sampled V-V interval values corresponding to the V-V intervals that occur most frequently in the distribution; comparing the HRDI to a specified HRDI threshold; and generating an indication of AF when the determined HRDI satisfies the HRDI threshold.

In example 20, the subject matter of one or both of examples 18 and 19 optionally includes arrhythmia detection circuitry configured to: determining a difference between the V-V intervals; determining a measurement of the V-V interval dispersion using the determined V-V interval difference; comparing the measurement result of the V-V interval dispersion with a specified dispersion threshold value; and generating an indication of AF when the determined measure of V-V dispersion satisfies a specified dispersion threshold.

Example 21 may include any portion of any portion or any portion of any one or more of examples 1-20 or may optionally be combined therewith to include subject matter that may include: means for performing any one or more of the functions of examples 1-20; or a machine-readable medium comprising instructions which, when executed by a machine, cause the machine to perform any one or more of the functions of examples 1-20.

These non-limiting examples may be combined in any permutation or combination.

The foregoing detailed description includes references to the accompanying drawings, which form a part hereof. The drawings show, by way of illustration, specific embodiments in which the invention may be practiced. These embodiments are also referred to herein as "examples. All publications, patents, and patent documents referred to herein are incorporated by reference in their entirety as if individually incorporated by reference. The use in the incorporated references should be considered supplementary to the present document if the usage between this document and those documents so incorporated by reference is inconsistent; for irreconcilable inconsistencies, the usage in this document controls.

In this document, the terms "a" or "an" are used, as is common in patent documents, to include one or more than one, independent of any other instances or usages of "at least one" or "one or more. Herein, the term "or" is used to refer to nonexclusive, such that "a or B" includes "a but not B," "B but not a" and "a and B," unless otherwise indicated. In the appended claims, the terms "including" and "in which" are used as the plain-english equivalents of the respective terms "comprising" and "wherein". Furthermore, in the following claims, the terms "comprising" and "including" are open-ended, that is, a system, apparatus, article, or process that includes elements in addition to those listed after such term in a claim is still considered to be within the scope of that claim. Furthermore, in the following claims, the terms "first," "second," and "third," etc. are used merely as labels, and are not intended to impose numerical requirements on their objects.

The method examples described herein may be at least partially machine or computer implemented. Some examples may include a computer-readable medium or machine-readable medium encoded with instructions operable to configure an electronic device to perform a method as described in the above examples. Embodiments of such methods may include code, such as microcode, assembly language code, higher level language code, or the like. Such code may include computer readable instructions for performing various methods. The code may form part of a computer program product. Further, the code can be tangibly stored on one or more volatile or non-volatile computer-readable media during execution or at other times. These computer-readable media may include, but are not limited to, hard disks, removable magnetic disks, removable optical disks (e.g., compact disks and digital video disks), magnetic cassettes, memory cards or sticks, Random Access Memories (RAMs), Read Only Memories (ROMs), and the like. In some examples, a carrier medium may carry code to implement the methods. The term "carrier medium" may be used to refer to the carrier wave on which the code is transmitted.

The above description is intended to be illustrative and not restrictive. For example, the above-described examples (or one or more aspects thereof) may be used in combination with each other. Other embodiments may be used, such as by one of ordinary skill in the art in view of the above description. The abstract is provided to allow the reader to quickly ascertain the nature of the technical disclosure. It is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims. Moreover, in the foregoing detailed description, various features may be grouped together to simplify the present disclosure. This should not be interpreted as intending that the disclosed features of no claim are essential to any claim. Rather, inventive subject matter may lie in less than all features of a particular disclosed embodiment. Thus the following claims are hereby incorporated into the detailed description, with each claim standing on its own as a separate embodiment. The scope of the invention should be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims

1. A device for detecting and identifying cardiac arrhythmias, comprising:

sensing circuitry configured to generate a sensed cardiac signal representative of cardiac activity of a subject; and

arrhythmia detection circuitry electrically coupled to the sensing circuitry and configured to apply Atrial Fibrillation (AF) detection criteria to the sensed cardiac signals, the AF detection criteria including:

primary AF detection criteria configured to detect an onset of AF in the sensed cardiac signal; and

a second AF detection standard configured to confirm the onset of AF in response to detection by a primary AF detection standard, wherein the primary AF detection standard has a higher sensitivity to AF detection than the second AF detection standard, and the second AF detection standard has a higher specificity to AF detection than the primary AF detection standard; and

a control circuit configured to trigger storage of sampled values of the sensed cardiac signal comprising an episode of AF when the episode of AF is detected by both the primary AF detection criteria and the second AF detection criteria.

2. The device of claim 1, wherein the arrhythmia detection circuit is configured to:

monitoring information corresponding to ventricular depolarization V-V intervals;

determining a V-V interval distribution using the sampled V-V interval values; and wherein the arrhythmia detection circuit, in accordance with the second AF detection criteria, is further configured to:

determining a heart rate density index, HRDI, as a portion of the sampled V-V interval values corresponding to the V-V intervals that occur most frequently in the distribution;

comparing the HRDI to a specified HRDI threshold; and

generating an indication of AF when the determined HRDI satisfies the HRDI threshold.

3. The device of claim 2, wherein the arrhythmia detection circuit, in accordance with the primary AF detection criteria, is configured to:

determining a heart rate mode as the heart rate corresponding to the V-V interval value having the most samples in the V-V interval distribution;

comparing the heart rate pattern to a specified heart rate pattern threshold; and

generating an indication of AF when the heart rate pattern meets the specified heart rate pattern threshold.

4. The device of claim 2, wherein the arrhythmia detection circuit, in accordance with the primary AF detection criteria, is configured to:

determining a difference between the V-V intervals;

determining a measurement of the V-V interval dispersion using the determined V-V interval difference;

comparing the measurement of the V-V interval dispersion with a specified dispersion threshold; and

generating an indication of AF when the determined measure of V-V dispersion satisfies the specified dispersion threshold.

5. The device of claim 1, comprising a physiological sensor circuit configured to generate a physiological signal comprising physiological information of the subject, wherein the arrhythmia detection circuit is configured to apply the second AF detection criteria to the sensed physiological signal to detect the onset of AF.

6. The device of claim 5, wherein the physiological sensor circuit comprises at least one of a heart sound sensor circuit and a pulmonary artery pressure sensor circuit.

7. The device of claim 1, wherein the arrhythmia detection circuit comprises a scoring module configured to determine a score associated with a correlation of a morphology of the sensed cardiac signal to a morphology of a template signal representing AF, and wherein the arrhythmia detection circuit is configured to apply at least one of the primary AF detection criteria and the second AF detection criteria to detect an onset of AF using the determined score.

8. The device of claim 1, further comprising a memory, wherein the memory comprises a first start buffer and a second event store buffer, and wherein the control circuitry is configured to: storing sample values of the sensed cardiac signal in the first start buffer and overwriting sample values previously stored in the first start buffer; and in response to AF detection by the primary AF detection criteria and the second AF detection criteria, initiate storage of sample values stored in the first start buffer in the second event storage buffer and subsequent sample values of the sensed cardiac signal in the second event storage buffer.

9. The device of claim 1, further comprising a memory, wherein the control circuit is configured to store the sampled values of the cardiac signal in the memory in response to detecting the onset of AF using the primary AF detection criteria, wherein the arrhythmia detection circuit is configured to detect the onset of AF using the stored sampled values of the cardiac signal using the second AF detection criteria and generate an indication of detection of AF, and wherein the control circuit is further configured to continue storing the sampled values of the cardiac signal in response to the generated indication of AF.

10. The apparatus of claim 1, further comprising a memory,

wherein the control circuit is configured to store sample values of a first segment of the cardiac signal in the memory for a first duration in response to the arrhythmia detection circuit detecting an onset of AF using the primary AF detection criteria,

wherein the arrhythmia detection circuit is configured to apply the second AF detection criterion to stored values of a first cardiac signal segment,

wherein the control circuit is configured to store sample values of one or more cardiac signal segments having a second duration that is less than the first duration when AF is detected by both the primary AF detection criterion and the second AF detection criterion,

wherein the arrhythmia detection circuit is configured to apply the second AF detection criterion to one or more cardiac signal segments having the second duration, and

wherein the control circuit is configured to continue storing sample values of the cardiac signal segments of the second duration while the arrhythmia detection circuit continues to detect AF in one or more cardiac signal segments of the second duration.

11. The device of any of claims 1-10, wherein the control circuitry is configured to determine that sample values of a segment of the cardiac signals have not been stored for a specified period of time, and in response to the determination, initiate storage of sample values of a segment of the cardiac signals including an episode of AF in memory when the episode of AF is detected only by the primary AF detection criteria.

12. The apparatus of claim 11, comprising: a therapy circuit configured to be coupled to the electrode to provide anti-arrhythmic cardiac therapy to the subject; and control circuitry configured to initiate delivery of anti-arrhythmic therapy in response to the generated indication of AF.

13. The apparatus of claim 11, wherein the sensing circuit is configured to be coupled to at least one of: an implantable electrode configured to be placed in a heart chamber or a subcutaneous implantable electrode configured to sense cardiac signals without direct cardiac contact with the subject.

14. The apparatus of claim 11, wherein the sensing circuit and the arrhythmia detection circuit are included in a wearable device or a handheld device.

15. A device for detecting and identifying cardiac arrhythmias, comprising:

sensing circuitry configured to generate a sensed cardiac signal representative of cardiac activity of a subject;

an arrhythmia detection circuit configured to detect an onset of Atrial Fibrillation (AF) in the sensed cardiac signal; and

a control circuit configured to:

initiating sensing the cardiac signal during a specified detection duration as an AF detection window to detect an onset of AF in the sensed cardiac signal using primary AF detection criteria;

initiating sensing the cardiac signal during a first AF detection window and a consecutive second AF detection window in order to confirm the onset of AF using a second AF detection criterion, wherein the primary AF detection criterion has a higher sensitivity to AF detection than the second AF detection criterion and the second AF detection criterion has a higher specificity to AF detection than the primary AF detection criterion; and

trigger storing of sampled values of the sensed cardiac signal in response to detecting AF during both the first AF detection window and the second AF detection window.